We propose to conduct a prospective cohort study to address three primary hypotheses: oral contraceptive use, particularly longer term use before the first pregnancy, increases the risk of breast cancer; physical activity during young adulthood reduces the risk of breast cancer; and alcohol intake before age 30 years increases the risk of this malignancy. To accomplish this we will enroll 125,000 registered nurse 25 to 42 years of age and collect information on these exposures and known risk factors for breast cancer by means of a mailed questionnaire accompanied by a color booklet to facilitate the reporting of specific brands and type of oral contraceptives. Follow-up questionnaire will be mailed at two-year intervals to update information on exposure status and to ascertain incident nonfatal diagnoses of cancer, which will be further documented by review of medical records. Deaths among nonrespondents will be ascertained by use of the National Death Index. During four years of follow-up we anticipate a sufficient number of breast cancer cases to detect relative risks of 1.7 for women using oral contraceptives four or more years before their first full- term pregnancy, which is similar to the magnitude of the association found in several recent case-control studies. This proposed study will be closely patterned after the ongoing Nurses Health Study but will address issues that are not possible to examine within the present study since the youngest cohort members are now 42 years of age. This new cohort will take advantage of our considerable experience and resources developed in the conduct of the Nurses Health Study and Health Professionals Follow-up Study. this includes not only the expertise of the investigators, but also the skills of our programmers and other support staff, computing and data processing facilities, and an extensive body of management and analysis software especially designed for large cohort studies. Through this effort, we hope to identify avoidable risk factors for breast cancer, a common disease without established practical means of prevention.